摘要
短期电力负荷预测方法通过分析历史负荷数据提取负荷参数特征,因缺乏对历史负荷离群数据的有效处理,负荷预测精度较差。因此,提出基于数据挖掘的短期电力负荷预测方法。首先,结合数据挖掘方法检测离群负荷数据,提高负荷序列的完整度;其次,对负荷数据进行数据清洗及归一化处理;最后,参考负荷序列,对短期负荷进行预测。实验结果表明,利用该方法预测电力负荷时,预测曲线与实际负荷曲线间的拟合程度较高,取得理想的负荷预测精度。
The short-term power load forecasting method extracts load parameter features by analyzing historical load data,but due to the lack of effective processing of historical load outlier data,the accuracy of load forecasting is poor.Therefore,a short-term power load forecasting method based on data mining is proposed.Firstly,combining data mining methods to detect outlier load data and improve the integrity of load sequences.Secondly,clean and normalize the load data.Finally,refer to the load sequence to predict short-term loads.The experimental results show that when using this method to predict power loads,the fitting degree between the predicted curve and the actual load curve is high,achieving ideal load prediction accuracy.
作者
王季琴
李志霞
WANG Jiqin;LI Zhixia(School of Electrical Engineering,Henan Mechanical and Electrical Vocational College,Zhengzhou Henan 451191,China)
出处
《信息与电脑》
2023年第15期58-60,共3页
Information & Computer
关键词
数据挖掘
聚类分析
负荷预测
数据清洗
离群数据
data mining
cluster analysis
load forecasting
data cleaning
outlier data
